A. Field of the Invention
The invention relates generally to computer graphics and image processing, and more particularly to methods and apparatus for processing input gray level values with dither matrices to generate output color representative of an output image.
B. Description of the Related Art
Dithering is a graphics technology for selecting an approximate color from a mixture of other colors by a computer program when a screen is unable to display the color required by the image. That is, pixel data having high bits (N bits) are converted to pixel data having low bits (M bits).
FIG. 1 is a block diagram showing a conventional apparatus for performing a dithering process. As shown in FIG. 1, a dither value generator 101 generates a dither reference value to a dithering unit 102 in response to a pixel address. Then, an original pixel color value (N bits) is input to the dithering unit 102 for generating a dithered value (M bits) by performing a dithering process in response to the dither reference value.
FIG. 2 is a block diagram showing a conventional apparatus for performing an inversely dithering process. As shown in FIG. 2, the dither value generator 201 generates a dither reference value in response to the pixel address. The dithered value (M bits) of the pixel is input to the inversely dithering unit 202, which generates a recovered value (N bits) by performing an inversely dithering process in response to the dither reference value.
However, the dithering process will result in “fixed pattern effect” because the dither value generator 201 of the prior art can select a dither reference value from a dither matrix Dn merely based on a pixel address. For example, for a pixel that is located at (x, y) on the screen and has a color value of I (“pixel data” hereinafter), the corresponding row and column of the pixel's dither reference value are calculated as:                i=(x modulo n)        j=(y modulo n).        
The dither reference values are selected from a dither matrix in accordance with the values of (i, j). By comparing the pixel color value I with the dither reference value Dn(i, j), a value I′ can be defined in accordance with the following formulas. The highest M bits of the value I′ is used to represent the dithered value of the pixel data.                I′=I−(I modulo 2N-M), if I<Dn(i, j); or        I′=I−(I modulo 2N−M)+2N−M, if I>Dn (i, j).        
Since the image generates a “fixed pattern effect” frequently after the dither matrix Dn is used to perform dithering process a number of times, even though a relatively large dither matrix is employed, the “fixed pattern effect” can only be alleviated without being resolved completely.
Moreover, since the least significant bits (LSB) of the color value are abandoned in the conventional dithering techniques, a random value must be used to perform the inversely dithering process. Therefore, after performing dithering process a number of times, it may result in a significant “color deviation effect”.